-------------------------------------------------------------------------------help parmest_ci_opts(Roger Newson) -------------------------------------------------------------------------------

Confidemce-interval options forparmestandparmby

Syntax

optionsDescription -------------------------------------------------------------------------eformExponentiate estimates and confidence limitsdof(numscalar)Scalar degrees of freedom for calculating confidence limitslevel(numlist)Confidence level(s) for calculating confidence limitsclnumber(numbering_rule)Numbering rule for naming confidence limit variablesmcompare(method)Multiple-comparison methodmcomci(method)Multiple-comparison method for confidence limits onlybmatrix(matrix_expression)Matrix from which parameter estimates will be extractedvmatrix(matrix_expression)Matrix from which parameter variances will be extracteddfmatrix(matrix_expression)Matrix from which parameter degrees of freedom will be extracted -------------------------------------------------------------------------where

numscalaris# |

scalar_nameand

numbering_ruleis

level|rankand

methodis

noadjust|bonferroni|sidak

DescriptionThese options allow the user to change the selection of confidence limits and

P-values in the output dataset (or resultsset) created byparmestorparmby.

Options

eformspecifies that the estimates and confidence limits in the output dataset are to be exponentiated, and the standard errors multiplied by the exponentiated estimates. This option is usually used if the estimated parameters were calculated on a log scale, as is done bylogitandlogisticwith odds ratios, byglmandbinregwith risk ratios, bystcoxwith hazard ratios, or byregresswith geometric mean ratios. Note that, if the user wants exponentiated confidence intervals in the output dataset, then theeformoption must be specified forparmbyorparmest, whether or not theeformor equivalent option was specified for the estimation command.

dof(numscalar)specifies the scalar degrees of freedom fort-distribution-based confidence limits andP-values. Ifdof()is positive, then confidence limits andP-values for all parameters are calculated using thet-distribution withdof()degrees of freedom. Ifdof()is zero, then confidence limits are calculated using the standard Normal distribution. Ifdof()is absent (or missing or negative), then confidence limits are calculated from the standard Normal ort-distribution, as follows. If thedfmatrix()option specifies a valid degrees of freedom matrix (see below), then the degrees of freedom are extracted from the specified matrix. Otherwise, if there is a non-missing scalar estimation resulte(df_r), then the degrees of freedom for all parameters is set to the value ofe(df_r). Otherwise, the confidence limits andP-values are calculated using the standard Normal distribution.

level(numlist)specifies the confidence levels, in percent, for the confidence limit variables created in the output dataset. These levels do not have to be integers, but must be at least 0 and strictly less than 100. For each levelxx,parmestandparmbycalculate a lowerxxpercent confidence limit variable with a default name of formminxxand an upperxxpercent confidence level with a default name of formmaxxx. The numbering of the confidence limit variable names can be changed using theclnumberoption (see below), and the names of the confidence limits can be changed using therenameoption (seeparmest_varmod_opts). The default islevel(95), or another single number set byset level. Note that thelevel()option used byparmbyorparmestis not affected by anylevel()option specified for the estimation command. (See[U] 20 Estimationand postestimation commands.)

clnumber(numbering_rule)specifies the rule used to number the names of the confidence limit variables created in the output dataset. Thenumbering_rulemay belevelorrank, and is set in default tolevelif theclnumber()option is not specified. For each confidence levelxxspecified by thelevels()option,parmestandparmbycalculate a lowerxxpercent confidence limit with the default nameminyy, and an upperxxpercent confidence limit with the default namemaxyy, where the numberyydepends on the confidence levelxxaccording to a rule specified by thenumbering_ruleof theclnumber()option. If thenumbering_ruleisrank, then the numberyyis the rank, in ascending order, of the confidence levelxxin the set of confidence levels specified by thelevels()option. For instance, if the user specifieslevels(90 95 99) clnumber(rank), then the 90 percent confidence limits are namedmin1andmax1, the 95 percent confidence limits are namedmin2andmax2, and the 99 percent confidence limits are namedmin3andmax3. If the numbering rule islevel(the default), then the numberyyis equal to the confidence levelxx. For instance, if the user specifieslevels(90 95 99) clnumber(lewel), then the 90 percent confidence limits are namedmin90andmax90, the 95 percent confidence limits aremin95andmax95, and the 99 percent confidence limits aremin99andmax99. If the confidence levelxxcontains a decimal point, then the decimal point is replaced with "_" in the variable namesminxxandmaxxx. If the confidence levelxxcontains "e-" (because of very small e-format confidence levels), then the "e-" is replaced with "em" in the variable namesminxxandmaxxx. Therefore, if the user specifieslevel(95 99.99)clnumber(level), then the output dataset contains 95% lower and upper confidence limits in variablesmin95andmax95, and 99.99% lower and upper confidence limits in variablesmin99_99andmax99_99. The optionclnumber(rank)is useful if the confidence levels contain many numbers after the decimal point, which may be the case if the user specifies Bonferroni-corrected or Sidak-corrected confidence limits.

mcompare(method)specifies a multiple-comparison method used to adjust the generated confidence limits andP-values. This method may benoadjust(the default, indicating no adjustment),bonferroni(indicating the Bonferroni adjustment), andsidak(indicating the Sidak adjustment). The adjustments, if requested, are calculated for the total number of parameters estimated (in the case ofparmest), or for the number of parameters estimated for the by-group (in the case ofparmby). For this reason, themcompare()andmcompci()options are not used very often withparmestandparmby, and are more likely to be used withparmcipin a derived resultsset, containing subsets of parameters from multiple models.

mcomci(method)specifies a multiple-comparison method used to adjust the generated confidence limits only, and not the generatedP-values. If the user wants to generate adjustedP-values without adjusting the confidence limits, or to generate adjustedP-values using a different method from the one used for adjusting the confidence limits, then the user is advised to use theqqvaluepackage, which can be downloaded from SSC. Note that themcompci()andmcompare()options do not affect the names of the generated variables containing the confidence limits, only their values.

bmatrix(matrix_expression)specifies the matrix from which the parameter estimates will be extracted. If not set by the user, then it is set toe(b)for most estimation commands, or toe(b_mi)if the most recent estimation command ismi estimate, or toe(est)if the most recent estimation command is one of the superseded Stata 8 survey commandssvymean,svyratioorsvytotal, and the command was specified with theavailableoption instead of thecompleteoption. The matrix specified must have one row, and one column per estimated parameter. The column names and equations of the matrix are used as the source for the parameter names and equations in the output dataset.

vmatrix(matrix_expression)specifies the matrix from which the parameter variances will be extracted. If not set by the user, then it is set toe(V)for most estimation commands, or toe(V_mi)if the most recent estimation command ismi estimate, or toe(V_db)if the most recent estimation command is one of the superseded Stata 8 survey commandssvymean,svyratioorsvytotal, and the command was specified with theavailableoption instead of thecompleteoption. The matrix specified must have as many columns as the matrix specified bybmatrix(), and must either have one row (from which the variances will then be extracted), or have as many rows as columns (in which case the variances will be extracted from the diagonal).

dfmatrix(matrix_expression)specifies the matrix from which the parameter degrees of freedom will be extracted, if nodof()option has been specified by the user. If neitherdof()nordfmatrix()has been specified by the user, then, for most estimation commands, the degrees of freedom for all parameters are extracted from the scalare(df_r)if this result is not missing, and the standard Normal distribution is used otherwise. However,dfmatrix()is set in default toe(df_mi)if the most recent estimation command ismiestimate, or toe(_N_psu)-e(_N_str)if the most recent estimation command is one of the superseded Stata 8 survey commandssvymean,svyratioorsvytotal, and the command was specified with theavailableoption instead of thecompleteoption. The matrix specified must have one row, and must have either one column (from which degrees of freedom will be extracted for all parameters), or as many columns as the matrix specified bybmatrix(). Note thatdfmatrix()is ignored if the user specifiesdof().

Selection of distribution and degrees of freedom

parmestandparmbycalculate confidence intervals andP-values from the parameter estimates and standard errors, using either the standard Normal distribution or thet-distribution for all parameters. If thet-distribution is used, then the degrees of freedom may or may not be the same for all parameters. The distribution, and degrees of freedom, are selected as follows:1. By first preference, the

dof()option is used, if specified by the user.2. By second preference, the

dfmatrix()option is used, if specified either by the user or by default.3. By third preference, the degrees of freedom are specified by the scalar estimation result

e(df_r), if that result is present.4. If none of the above possibilities are available, then the standard Normal distribution is used.

Note that the user can force the use of the standard Normal distribution by specifying

dof(0), or force the use ofe(df_r)(if present) by specifyingdof(e(df_r)).If the

t-distribution is used, then the degrees of freedom for each parameter are stored in the output dataset in the variabledof, and thet-test statistics are stored in the variablet. If the standard Normal distribution is used, then the output variabledofis not created, and thez-test statistics are stored in the output variablez.

AuthorRoger Newson, Imperial College London, UK. Email: r.newson@imperial.ac.uk

Also seeManual:

[U] 20 Estimation and postestimation commandsHelp:

[U] 20 Estimation and postestimation commandsparmest,parmby,parmest_outdest_opts,parmest_varadd_opts,parmest_varmod_opts,parmby_only_opts,parmest_resultssetsqqvalueif installed